diy solar

diy solar

Co-worker says Im wrong. Maybe you can tell me if I'm crazy.....

MorganCarey

New Member
Joined
Jul 19, 2020
Messages
49
OK, so I don't have a HUGE amount of experience building battery banks but this is a pretty basic question as it relates to SLA batteries.

For context, I have built a DIY LiFePO4 24V battery. I top balanced all cells in parallel as per all diy writeups. The concept makes sense to me. I can see the difference in my BMS stats for each cell if any get out of balance.

The simple question is... is this balancing required for SLA batteries that are to be wired in series?

If you're interested in the debate keep reading:

I have a new position at work in the telecom sector. I was previously a line-man of sorts, but I've moved into a new role that finds me responsible for switching equipment maintenance in the telecom network. Part of this is testing and maintenance of battery strings. We use SLA batteries on our terminals. They are 48V strings made up of 12V SLA batteries wired in series.

These batteries are charged and topped up by rectifiers that run off mains. There is NO battery balancing mechanism in play, nor any conditioning cycles. These are straight 48V rectifiers. Target voltage is approximately 52V.

The batteries will on occasion be discharged to below 50% during power outages. I believe critical equipment shutdown happens around 42 volts. They absolutely do hit this depth of discharge easily a few times a year as these machine are located in rural areas where the hydro grid is susceptible to storm damage with extended outages.

OK, if you're still with me, thanks for sticking around...

The debate:

Recently I identified an issue with a battery in an existing string. It had shorted itself. Policy now as the budgets shrink is to replace the single battery in the string.

Again, I'm new to the position, but basically I was going to bring all the batteries back to the office and balance charge them along with the new cell over night. My mentor said I'm crazy and that's not required. I asked, well should we at least charge the replacement battery first before adding it back to the string in production (it had been sitting on a shelf for who knows how long at our supplier). His reply was basically no, and that the batteries will self balance while they are wired in series and hooked up to the rectifier.

This doesn't sound right to me. If they were Lithium batteries this would definitely be wrong, but these are SLA and I don't have a ton of experience here with SLAs wired in series.

What say you guys?
 
The 3 old batteries are already going to bring the new battery's performance, you should definately pre-charge the batteries to the same state of charge.

It sounds like 1: the boss doesn't know much about batteries and 2: he's being cheap and would rather buy new batteries later than pay you for the time to do the job right the first time.
 
His reply was basically no, and that the batteries will self balance while they are wired in series and hooked up to the rectifier.

I think lead-acid batteries will self-diverge.
The one that is lower will degrade faster, resulting in it getting even lower.

Unless the series connected batteries are charged high enough to equalize. Those completely full sit there outgassing while passing current to the low one so it can get full. In the case of FLA, replace lost water. In the case of AGM, catalytic recombination should turn the released H2 & O2 back into H2O, if not done too rapidly.

But no reason to start off on the wrong foot. Bring all batteries up until they have the same fully-charged resting voltage before using in series. Series connected equalizing is less preferred than charging each battery or cell individually.

If you have them at same voltage less than fully charged, when fully charged they will diverge due to differing capacity. I just observed this assembling what had been 2 strings of 12 (8V) batteries into 3 strings of 8.

How many 48V strings do you have in parallel? If many, how about setting these aside to be held at float individually, use as spares later?
 
If you take the one bad battery out to just replace it. The new battery will always perform to the age of the other 3 in the string. Basically making a brand new battery act the same age as the the 3 older ones.

Typical VRLA batteries used in telecom are good quality. Some you can equalize according to manufacturer instructions, some say not to equalize.

My experience was to replace the whole string if batteries where more than 1.5 years old.
If less than 1.5 years old, take out the bad battery. Top off the other three good batteries individually, till they where all within .01 volts of each other. Then place them back in service with the new battery.
The new battery would take on the characteritics off the three older batteries as far as capability.

Edit: If you can't get the older batteries to hold a charge that is within the .01 volts, then we would replace other batteries in the string to achieve this .01 volt tolerance. See the manufacturer instructions for your particular Telecom battery, as manufacturers differ on voltage spread between batteries.
 
Last edited:
OK, so I don't have a HUGE amount of experience building battery banks but this is a pretty basic question as it relates to SLA batteries.

For context, I have built a DIY LiFePO4 24V battery. I top balanced all cells in parallel as per all diy writeups. The concept makes sense to me. I can see the difference in my BMS stats for each cell if any get out of balance.

The simple question is... is this balancing required for SLA batteries that are to be wired in series?

If you're interested in the debate keep reading:

I have a new position at work in the telecom sector. I was previously a line-man of sorts, but I've moved into a new role that finds me responsible for switching equipment maintenance in the telecom network. Part of this is testing and maintenance of battery strings. We use SLA batteries on our terminals. They are 48V strings made up of 12V SLA batteries wired in series.

These batteries are charged and topped up by rectifiers that run off mains. There is NO battery balancing mechanism in play, nor any conditioning cycles. These are straight 48V rectifiers. Target voltage is approximately 52V.

The batteries will on occasion be discharged to below 50% during power outages. I believe critical equipment shutdown happens around 42 volts. They absolutely do hit this depth of discharge easily a few times a year as these machine are located in rural areas where the hydro grid is susceptible to storm damage with extended outages.

OK, if you're still with me, thanks for sticking around...

The debate:

Recently I identified an issue with a battery in an existing string. It had shorted itself. Policy now as the budgets shrink is to replace the single battery in the string.

Again, I'm new to the position, but basically I was going to bring all the batteries back to the office and balance charge them along with the new cell over night. My mentor said I'm crazy and that's not required. I asked, well should we at least charge the replacement battery first before adding it back to the string in production (it had been sitting on a shelf for who knows how long at our supplier). His reply was basically no, and that the batteries will self balance while they are wired in series and hooked up to the rectifier.

This doesn't sound right to me. If they were Lithium batteries this would definitely be wrong, but these are SLA and I don't have a ton of experience here with SLAs wired in series.

What say you guys?

Replacement batteries should already be fully charged.

You're probably not allowed to take a bunch of batteries out of service at any given time.

Do what your boss says.
 
If you take the one bad battery out to just replace it. The new battery will always perform to the age of the other 3 in the string. Basically making a brand new battery act the same age as the the 3 older ones.

Typical VRLA batteries used in telecom are good quality. Some you can equalize according to manufacturer instructions, some say not to equalize.

My experience was to replace the whole string if batteries where more than 1.5 years old.
If less than 1.5 years old, take out the bad battery. Top off the other three good batteries individually, till they where all within .01 volts of each other. Then place them back in service with the new battery.
The new battery would take on the characteritics off the three older batteries as far as capability.

Edit: If you can't get the older batteries to hold a charge that is within the .01 volts, then we would replace other batteries in the string to achieve this .01 volt tolerance. See the manufacturer instructions for your particular Telecom battery, as manufacturers differ on voltage spread between batteries.
Thanks for this reply. Much appreciated and sheds a bit of light on this for me. Thank you.
 
I use a balancer with my ex-telco SLA batteries. Enersys SBS-190F.
But I charged them all first before putting them in series.
 
I think lead-acid batteries will self-diverge.
The one that is lower will degrade faster, resulting in it getting even lower.

Unless the series connected batteries are charged high enough to equalize. Those completely full sit there outgassing while passing current to the low one so it can get full. In the case of FLA, replace lost water. In the case of AGM, catalytic recombination should turn the released H2 & O2 back into H2O, if not done too rapidly.

But no reason to start off on the wrong foot. Bring all batteries up until they have the same fully-charged resting voltage before using in series. Series connected equalizing is less preferred than charging each battery or cell individually.

If you have them at same voltage less than fully charged, when fully charged they will diverge due to differing capacity. I just observed this assembling what had been 2 strings of 12 (8V) batteries into 3 strings of 8.

How many 48V strings do you have in parallel? If many, how about setting these aside to be held at float individually, use as spares later?
I agree with you about not starting off on the wrong foot. However, I guess in many cases it will all come out in the wash.

In the machines these run, most only have a single string. The machines were designed for two strings, but as subscriber base drops, the company has started to go to single strings in many machines.
 
I spent the last 40 years in the telecommunications industry. Replacing one bad battery will buy you a year and then you’ll have 4 bad batteries…plus a good chance of an outage. My role has been in public safety communications and outages can cost lives. Having the luxury of being old and opinionated, I will say your boss is wrong. His approach to saving money will eventually cost more than he saved. Remembering when I was in your position however, I would refer to rule number one which states the boss is always right. Rule number two states when the boss is wrong, refer to rule number one. Cut him some slack though because he probably knows this but has budget pressures to consider and his hands are tied. I’ve been on both sides and management doesn’t have the luxury of looking past the end of the next fiscal year.
 
Consider also that the amount you earn per hour is nowhere near the amount of money the company has to pay to keep you employed. An employee paid $40 per hour could easily be costing the company $80-$100/hour.

With that in mind do as the boss says, if you waste several hours hauling batteries around to try to save them and you spend several hundred dollars of the companies money just to be right and make your boss wrong then don't be surprised if his self preservation kicks in and you find yourself back down as a lineman.

If there was a way to do this properly AND stay within budget he would probably get you to do it that way. He's working with the constraints he has on his job so screwing him over to save a battery isn't a great life choice. You won't win by doing this.

If you don't like the way the company is doing things find another company that is more to your liking.

If one of my employees did this after I told them not to I'd fire them.
 
Hey Everyone. Thanks a lot for all your replies.

To clarify a few things....

The decision to only change one battery is not in my "boss's" purview. That's a decision made by bean counters at corporate. If it was up to him, he'd say go for it and replace all of them. The time spent on this wouldn't really be an issue for him either I'd imagine. When I eventually take the reigns, I might consider making the decision on my own to spend the time to do this the most 'optimal way' as it doesn't really seem to be a hard and fast rule one way or another. Theoretically, I could strategically pick a couple of days where it's going to be good weather and low risk of mains outage and pull the batteries to balance.

My 'mentor' is a very senior facilities and maintenance tech that I will be replacing when he retires in a couple of years. He's a great guy and hasn't lead me astray yet. Going by some of the knowledgeable replies here - his advice, is not necessarily technically wrong, even if it's not optimal. Though his advice gave me pause given my experience with lifepo4 and therefore I came here, to a pool of very knowledgeable battery folks for some opinions.

Glad I did.

Thanks again everyone!
 
You don't have to disconnect to balance. You could connect the string to it is there for backup, and charge each 12V battery in situ.
(I tried to do that in my system and Sunny Island I think let the others draw down; your system may not behave that way.)

If the old string is still there, with one bad battery in series, the other three are probably fully charged, even overcharged. So fully charge the new one before swapping in.
 
Propose this: when the next battery needs to be replaced, ask if you can replace all 4 with new batteries. Keep the other 3 on a maintainer. When the next battery goes bad, you have a used battery you can replace it with.
 
Back
Top